In the present study, we tested the assumption that structural similarity overcomes surface similarity in the retrieval of past events, by observing whether structural similarity alone is a better cue than surface similarity alone. To do so, in three story-recall experiments, we provided the participants with multiple source stories and then with a target cue story. This target cue only shared surface similarity with one source story, and structural similarity with another source story. In Experiment 1A, a Superficially Similar Disanalog source story (SSD) and a Superficially Dissimilar Analog source story (SDA) were presented among Superficially Dissimilar Disanalog source stories (SDDs). A soundness rating task was used in Experiment 1B to control the absence of structural similarity among the SSDs presented in Experiment 1A. In Experiment 2, the number of SSDs was increased in the aim to reproduce more ecological conditions. In Experiment 3, a filler task was introduced and supplementary source stories were presented in order to make the study more similar to previous story-recall paradigms. The results of the three story-recall experiments support the dominance of structural over surface similarities in analogical retrieval. The role of a structurally-based access regarding the retrieval of Superficially Similar Analogs (SSAs) and SDAs is discussed, as well as the factors underlying the rare occurrence of SDAs retrievals in previous experiments.

Analogies are fundamental to our minds as they allow us to interpret incoming experiences, which are always, strictly speaking, new, in light of more familiar situations. These comparisons are particularly useful to guide comprehension in that they are not solely based on the superficial appearance of situations, but on deeper commonalities making a given situation essentially similar to another one. As such, they are crucial to guide the resolution of new problems by transferring solution procedures associated with analog problems solved in the past.Certain analogies may be implemented in a relatively straightforward way. A new situation is conceptualized in the same way as a previously encountered situation, and this common conceptualization leads to the two situations being brought together. For instance, one may easily assimilate a royal crown and a roman laurel wreath despite their distinct appearances, as they are first and foremostly seen as symbols of authority. The objects from this comparison may not have to be perceived from a new and unusual perspective. However, there are analogies that may precisely depend on the ability to change our point of view about a given object or situation. The famous analogy drawn by Archimedes to verify whether the king's crown was entirely made out of gold illustrates this necessity (Goswami 1992;

Ce chapitre propose un regard croisé sur la littérature scientifique dans le domaine de l’analogie et de la flexibilité cognitive. La flexibilité est définie comme un mécanisme essentiel à la mise en œuvre de certaines analogies. Des pistes éducatives sont dégagées à la lumière des travaux destinés à stimuler la flexibilité pour favoriser la mise en œuvre d’analogies rarement perçues par les élèves.